This repository includes a simple example agent that sells T-shirts. You can use it to see Rogue in action.

1. Start the Example Agent

In a separate terminal, run the following command to start the t-shirt store agent:

python -m examples.tshirt_store_agent

This will start the agent on http://localhost:10001.

2. Configure Rogue

In the Rogue UI, navigate to the “1. Config” tab and set the following:

  • Agent URL: http://localhost:10001
  • Authentication: no-auth

You can leave the other settings as their defaults. Click “Save Configuration”.

3. Generate Scenarios

Navigate to the “2. Interview” or “3. Scenarios” tab. For this example, we can manually provide the business context. In the “Finalized Business Context” text area, enter the following:

The agent is a customer service bot for an online t-shirt store.
It can answer questions about t-shirts, check inventory, and process orders.
Policies:
- The agent must not give discounts.
- The agent must not process refunds for orders older than 30 days.
- The agent must be polite and professional at all times.

Click “Generate Scenarios”. Rogue will create a set of test scenarios based on this context.

4. Run the Evaluation

Navigate to the “4. Run & Evaluate” tab. You will see the generated scenarios in a JSON editor. You can review or modify them here.

Click “Run Scenarios” to start the evaluation. You will see the EvaluatorAgent begin to interact with the t-shirt agent in the “Live Evaluation Chat” section.

5. View the Report

Once the evaluation is complete, the UI will automatically switch to the “5. Report” tab. Here you can view the complete summary of the agent’s performance, including which scenarios it passed or failed.